JPH0536316A - Manufacture of superconductive wire rod of small-gage wire filament and extra small-gage wire filament - Google Patents

Abstract

PURPOSE: To facilitate the manufacture of a flexible superconducting wire having very many filaments. CONSTITUTION: In a method for manufacturing a superconducting body constituted of very many single core filament rods 10 having a circular cross section, the following process is included that plural single core filament rods 10, having a circular cross section, are incorporated into the inner side of a thin thickness hexagonal tube 16, and these plural tubes 16 are arranged into an extrusion vessel 18, and then hardened, extruded, and drawn into the desired dimension of a wire rod.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a superconducting wire made up of a large number of filaments each having a circular cross section and having a single core. More specifically, the present invention relates to a method for producing a fine multicore superconducting wire and a superconducting superconducting wire having high quality and lower cost than conventional ones.

[0002]

2. Description of the Related Art NbTi, a fine wire filament used in electromagnets of accelerators such as superconducting super colliders (SSC)
The superconducting wire is required to minimize the distortion of the magnetic field related to the particle orbit. The magnetization of the filaments increases these strains. This magnetization has a critical current density (Jc
), NbTi volume fraction and filament diameter. The volume fraction of NbTi is usually fixed due to the operating current and stability requirements of the electromagnet. Jc should be as large as possible in order to minimize the proportion of expensive NbTi. Therefore, in order to minimize filament magnetization, the filament diameter should be as small as possible. It has been shown that a filament having a diameter of 2 μm eliminates the need to correct the bending of the magnetic field in the dipole magnet, and as a result, the cost of the huge accelerator can be significantly reduced.

A significant problem in the mass production of this type of wire is the large number of 2 μm diameter filaments required to obtain the required volume fraction of NbTi. Such conductors require 40,000 to 70,000 filaments. The manufacture of these conductors typically requires multiple expensive extrusion steps.

The finer the filaments, the more likely they are to deform non-uniformly, resulting in filament necking or sausage. As a result, the obtained Jc is lowered. If the filament spacing is reduced or the strength of the material between the filaments is increased, the filament sausage problem is solved. However, if the filaments are too close together, electrical coupling will occur, necessitating the use of expensive correction coils. Interfilament matrix materials, such as copper-nickel and manganese-doped barriers, that replace current matrix materials must be developed to the point where they can be practically incorporated into ultrafine wire conductor designs. This modification of the matrix material has to be a minimal compromise on the electrical stability of the conductors currently obtained with high purity copper matrices.

As far as superconducting applications such as AC devices and accelerator magnets are concerned, microfilamentary conductors have advantages. The manufacture of such conductors has been responsive to these needs with varying stages of success. 1986 P. Dubots et al. Proposal "Use at 50 / 60Hz
NbTi Extra Fine Filament Wire "Low Temperature Technology Material RPReed
And AF Clark, Vol. 32, Plenum Press, New York, 1986, pp.747, their filament is 1
It was reported to be at least μm. However, the Jc of the wire was far from the value obtained for the fully utilized thick filament wire. 1986 Intermagnetics General Corp. (IGC) for SSC wire rod research and development
Manufactured an SSC inner grade wire with a 2.7 μm filament (CG King et al.'S paper “Production and Identification of NbTi Fine Wire Filament in Copper Matrix” Low Temperature Technology Materials RPReed and AF Clark
Volume 32, Plenum Press, New York, 1986,
pp.731 and "Prototype production of NbTi wire rod for ultra-fine wire filament for SSC" IEEE bulletin on Magnetics, Vol.
MAG-23, No. 2, pp. 1351, March 1987). This wire had a Jc of 2400 A / mm ^{2 at} 5T, but the filament quality was poor, the wire was brittle and the filament was magnetically coupled. The filament quality is low because the number of filaments used in the first extruding step of the multi-step extruding process used to manufacture this wire is relatively small, so to speak, a plurality of bundled element members are made. It was because. The brittle wire is a result of the dimensions of the conductor as well as the initial values of billet processing such as the cold work state of the alloy, the integrity of the barrier and its bond quality. Filament bonding is due to the small spacing between filaments in the final size conductor and the lack of a suitable matrix material between filaments.

Rather than incorporating a resistive interfilament matrix material that does not bond the filaments to complicate the design of the conductors, a single multifilamentary wire extrusion process produces conductors with a filament diameter of 5 μm and is critical. It was decided to concentrate on cost reduction by obtaining wires with higher current density. The lead wire of SSC manufactured today exceeds 3000A / mm ² at 5T, which is the best lead wire of the large synchrotron (Tevatron) at Fermi Lab.
Significant progress has been made in Jc when compared to 2200 A / mm ² .

Other methods and suggestions below, EW Co
llings's paper "Stabilization design considerations for fine filament Cu / NbTi composites" AF Clark and low temperature technology materials
RPReed, Volume 34, Plenum Press, New York, 1988, pp.867-878, E. Gregory et al.
0.5% by weight in "Unbonded conductor with 2.5 μm diameter filament designed for outer cables of dipole magnets" IEEE Bulletin on Magnetics, Vol. 25, No. 2, pp. 1926, 1989). Cu of manganese between filaments
It is added to the matrix. Electron spin flip scattering with manganese has been shown to eliminate filament binding up to filament sizes of less than 1.5 μm for filament spacings of approximately 0.3 μm. here,
Large numbers (22900 and 40000) of ultrafine hexagonal single core filaments are used and mass production of ultrafine wire conductors is not practical. The method of Gregory et al.
A Jc of A / mm ² was produced. This value is far from the range of 2750 to 3000 A / mm ² required by the SSC type superconductor. The "n" value reported as 13 indicates poor filament homogeneity and is unacceptable for the intended application.

Aiming at high Jc and n-values, the limited part of SSCs, future accelerator applications and the use of fine wire filament conductors in AC motors and generators will result in significant cost savings. There is still an opportunity to develop cost-effective products to do.

Although the above description has been particularly concerned with NbTi multifilamentary wires, it should be understood that generally a multifilamentary billet having a large number of corewires is required for the application of various superconductors. I have to. For example, there is a demand for multi-core billets of Nb ₃ Sn type conductors, and the same applies to others.

[0010]

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a method for manufacturing a superconducting wire at a lower cost than ever before. A further object of the present invention is to provide a method for producing a superconducting wire having a large number of fine wire filaments and extra fine wire filaments, in which defects caused by bonding of adjacent filaments are not a problem. Another object of the present invention is to provide a method suitable for mass production of a superconducting wire having a large number of fine wire filaments and extra fine wire filaments.

[0011]

Briefly, in accordance with the present invention, a method for producing a superconductor comprised of a large number of single filament filament rods of circular cross section, comprising a single core filament of circular cross section. A method is provided for collecting the rods and first assembling them inside a thin wall hex tube. A plurality of these tubes are placed in an extrusion container, solidified,
Extruded and drawn to desired final dimensions.

The above objects, features and advantages of the present invention are as follows:
The following more detailed description of the invention, taken in conjunction with the accompanying indivisible drawings, will show in part in detail and will in part be apparent.

[0013]

If the filament diameter is in the range of 2 to 3 μm, tens of thousands of filaments are required for the SSC-scale stranded wire. Practically, the large number of filaments can be combined into a stranded wire by bundling and extruding a plurality of times. A method of addressing the problems associated with the production of such superconducting wire is to utilize the commonly used triple extrusion technique with appropriate improvements to eliminate filament sausage and bonding.

The triple extrusion process is illustrated in FIG. In this method, a large number of extruded clad monofilament filament rods are repeatedly bundled twice in order to obtain the required number of filaments. The single-filament filament rod 10 is first extruded onto a hexagonal rod 20, which is then assembled into an extrusion container 22, which is extruded again to produce a hexagonal secondary element member 24. Processed, several secondary element members 24 are assembled in a second extrusion container 26, again extruded for the last time and then drawn to the required final dimensions.

Although this three-step process reduces overall yield and increases process time, this process is nevertheless more attractive with smaller filament sizes (less than 3 μm). .

The number of filaments in the second extrusion is an important problem because the proportion of filaments located on the outer ring of the secondary element member decreases as the number of filaments increases. In some of the early conductors, the number of filaments used in the multifilamentary element members used in combination was 7, and then 19 were tried. In these cases, 6/7 or 85% of the filament and 12/19
Or 63% are each in the outer row of the array.
These outer filaments, which have a large amount of copper on one side, are as unsupported as the inner filaments. As a result, the "n" value in these conductors is relatively low due to the extensive filament sausage. OST, a US manufacturer, tested 55 filaments, with 24/55 or 43.6% of the filaments unconstrained. Another manufacturer, Furukawa, tried 85 filaments with 30/55 or 35.3% of the filaments unconstrained. Lower "n" values than the single extruded multifilamentary conductors were also observed in these cases. In this way, it can be seen that using a large number of filaments, 253, in the second extrusion results in 42/253 or only 16.5% of the filaments being unconstrained. Interfilament matrix materials such as Cu-0.5 wt% Mn or CuNi are used to prevent electrical coupling of filaments with very close filament spacing that the final product will reach. When CuNi is used, CuNi
The hardness of CuNi helps maintain filament uniformity because its strength is closer to that of NbTi than Cu. This further presses the outer ring of the filament in the element member of the multifilamentary wire used by bundling a plurality of bundles, resulting in a more uniform bundle of secondary conductor wires in the final product. Stiffer container materials may be studied to further mechanically support the outer row of filaments in the second extrusion. The final assembly must have a pure copper core and a copper coating. This placement of high RRR copper provides the conductor with a significant degree of electrical and thermal stability.

It has been found that the Jc and n values obtained are high when the one-time bundling (hexagonal or circular cross-section) method is used. When a large number of filaments such as tens of thousands of filaments are included, it is very difficult to assemble the billet when the hexagonal element member or the element member having a circular cross section is used. Thin hexagonal rods tend to twist or bend, and circular cross-section element members tend to move out of a perfect hexagonal close-packed pattern, resulting in various shapes of stacking faults. It should also be noted that circular cross-section rods have better straightness than hexagonal rods and can be made more economically.

To enable high rate, high volume production of stranded wire for SSC or other large applications, the present invention provides a novel configuration of hexagonal cell billets and circular cell billets using circular cross section rod element members. Provide assembly technology. As shown schematically in FIG. 2, this method involves the hexagonal close packing (h) of a number of circular cross-section single core rods in a thin hexagonal tube.
cp) pattern, and arranging the plurality of hexagonal tubes inside the extrusion container. The straight wall of the hexagonal tube has a perfect hcp without stacking faults.
Retain symmetry. The material of the hexagonal tube must be thin enough that the local area ratio of the outer filaments is not reversed, and strong enough to withstand processing during billet assembly. Candidate materials for this are hard pure copper,
It is a diluted, supple alloy of copper or Ni such as CuNi or CuMn, or dispersion strengthened copper such as Cu-Al ₂ O ₃ . The stronger material supports the outer row of filaments more strongly, eliminating local filament sausage.

Furthermore, in each hexagonal cell, the complete h
Since the cp pattern is maintained, this method results in uniform placement of the filaments throughout the assembly. This is not the case in the final stage billet of the triple extrusion process described above.

The hexagonal cell method of the present invention can be used not only for large numbers of fine filaments (about 3 μm or less), but also for automated production of large quantities of billets. Automated methods utilize, for example, already filled hexagonal cells as a secondary assembly.

The assembling method of the present invention can be effectively utilized to assemble a number of thin single core filaments of circular cross section in the range of 3000 to about 100,000, especially about 6000 to about 50,000 such filaments. It is intended for use in building core filaments. Furthermore, the number of filament wire rods to be provided can be further increased by first bundling the multi-core wire rods in the hexagonal tube and bundling a large number of hexagonal tubes in the outer container. In particular, the outer container can be bundled after extrusion as well as in the final outer container above it. Such intermediate containers can be rounded and bundled into a final container, or extruded into a hexagonal shape and then bundled into a final outer container.

Using the method described above, this hexagonal cell method can be repeated to produce conductors with millions of filaments. Such conductors are described in U.S. Pat. No. 4,803,034 entitled "Superconductor with Controlled Laminar Pinning Center and Method of Making It".
It is used for wire rods with a pinning center introduced as described in No. 310.

The advantage of such a multifilamentary wire is that it reduces the number of steps for manufacturing such a multifilamentary wire. The reason is that the use of the hexagonal cell method reduces the number of extrusion steps required to produce such a conductor having a very large number of wires.

The finished billet is consolidated to remove all voids between tissues by techniques such as hot isostatic pressing (hip). During this strengthening process, the hcp structure is well retained because each group of monofilaments fits well into the hexagonal tube.

Generally, in practicing the method of the present invention, about 7 to 1000 thin wall hexagonal tubes are used to form the secondary hexagonal cell elements, and about 50 to about 500 such hexagonal tubes. Preferably tubes are used.

The various modern composite processing steps are common to both the prior art triple extrusion process and the hexagonal cell extrusion process described above. A significant variation in the process evolved from the initial attempts at producing 2 μm filaments in 1986 to produce compliant wire rods of final dimensions. The use of diffusion barriers was reasonably established. Hot isostatic pressing is used to strengthen the finished billet prior to extrusion. hip
Is also expected to increase binding at various interfaces. For example, in NbTi filaments, there are interfaces between the NbTi / barrier, barrier / matrix and bundled elements. The cold worked state of the final billet alloy is improved by heat treatment prior to extrusion which makes the billet suitable for processing into final sized wire. Generally, this temperature varies from about room temperature to about 600-700 ° C. Extrusion temperature and deformation result in a well bonded, high integrity composite.

The method of the present invention aims to maintain the quality of filaments to a high level, reduce the binding of filaments, and produce a long-length flexible conductor wire by the above-mentioned steps.

[0028]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT In the preferred embodiment of the present invention, first a plurality of
An assembly billet composed of NbTi filaments or rods 10 is used. Generally, NbTi rods can be inserted directly. However, instead, each NbTi rod may be covered with a barrier layer 12 and a Cu coating 14 as shown. These rods are incorporated inside a plurality of thin wall hex tubes 16 as shown in FIG.

The barrier 12 may typically be Nb, but may be other materials such as Ta, Ni, V, Zr, or combinations or alloys thereof.

Once assembled in a thin hex tube,
A number of such hexagonal tube secondary element members are located inside the extrusion container or case 18.

The entire assembly is then compressed, extruded and drawn to the desired dimensions.

Referring to FIG. 3, a single coated rod of circular cross section 10 is shown in perspective view showing a barrier layer 12 and a Cu coating layer 14.

FIG. 4 shows, in perspective view, the assembled billet prior to extrusion, in which a number of coated single core rods 10 of circular cross section are inserted into a number of thin wall hex tubes 16, all of which are in an outer extrusion container. It shows that it is in 18.

FIG. 5 shows that a large number of single core rods 10 are inserted into a large number of hexagonal tubes 16, all of which are outer extrusion containers 18
Figure 3 shows a cross section of the final wire, compressed, extruded and drawn, maintained in a nearly perfect hexagonal close packing pattern.

Further, starting with NbTi filaments, each rod can be an alloy of materials including, for example, NbTiHf, NbTiTa, NbTiZr, and the like.

Furthermore, any type of multifilamentary superconducting wire can be manufactured by utilizing the method described so far with respect to the manufacture of NbTi wire. For example, referring to FIG. 6, a method of using the method of the present invention to manufacture Nb ₃ Sn type multifilamentary wire is described. First, a single core filament of a copper clad niobium rod is manufactured. Niobium core 30
Is surrounded by a copper coating 32. Without extruding
These are packed in hexagonal thin copper tubes 34.
These are filled in the outer container 36. The center of the container is filled with copper 38. Barriers are commonly used, as is known in the tin nucleus method. After extrusion, the center may be punched and a mandrill inserted in the center. Then, tin is inserted in place of the mandrel, and a final heat treatment is performed to diffuse tin into the niobium stranded wire to form an Nb ₃ Sn multifilamentary wire.

Although the production of such Nb ₃ Sn type superconductor has been described in connection with the tin nucleus method, such a multifilamentary wire can be produced by another method.

To date, thin walled hexagonal tubes have been shown to be uniformly packed with each uniquely coated rod. However, the hexagonal tube arranged near the outer circumference of the last container does not have to be completely packed with single-core filaments. FIG. 7 shows a typical outer hex tube 40 with a unique coated single core filament wire 42 that occupies only a portion of the tube. The lower portion of the tube is shown as filled with copper material 44. In this case, this special tube is used to form the outer ring of hexagonal tubes in the container 36 shown in FIG. Then, when extruded and finally heat treated, the copper of the outer portion of the hexagonal tube fuses with the copper of the matrix material so that it is closer to the wire of circular cross section in the outer container. This allows
The appearance is similar to that of a conventional type of wire with the wire in a circular ring in the container.

Alternatively, as shown in FIG.
When the outer container 50 is packed with a unique hexagonal tube 52, the unique wire 54 can be packed in the annular space around the hexagonal tube. The wire may be a unique copper wire, a copper clad wire, or NbTi, packed with single core filaments of such wire.

As mentioned above, the advantage provided by the method of the present invention is that a large number of rods of circular cross section are assembled into a thin hexagonal tube in a hexagonal close packing pattern to form a secondary element member. , When assembled together in the extrusion container along with a number of these other secondary elements, a nearly complete h
By holding the cp pattern up to the final wire product that has been compressed, extruded and drawn.

The inherent advantages of this method are the ability to condition a large number of single core filaments, ie up to 100,000 single core filaments or more, and still maintain a near perfect hcp pattern for necking and The aim is to eliminate the loss of current density due to the coupling effect, which results in an excellent end product with a very large number of evenly spaced and non-bonded single-core filaments.

The use of a secondary element member of a hexagonal tube is very important in this method, such a secondary element member of a hexagonal tube being of the type described in the present invention. This is a new achievement in a right that has not been used until now for the production of stranded superconducting wire.

An evaluation of sample billets assembled according to the method of the present invention in comparison with other methods is shown below. The assembly of the SSC filaments outside of about 4200 filaments using a unit of hexagonal single core filaments is usually accomplished by using a unit of bundled single core filaments of circular cross section. Cell method 3
It takes 4 times longer. The filament number savings is more dramatic, with 8000 filaments (eg 6-2.5 μm SSC
Type conductor)). Here, a complete matrix arrangement can be completed by assembly by the hexagonal cell method in a fraction of the time required to bundle a large number of filaments at one time to obtain a complete array.

The hexagonal cell secondary assembly method provides a suitable method for bundling rods of small diameter single core filaments as small as 0.049 inches in diameter; this is approximately 40,000 for a typical 12 inch billet. It is a filament and still achieves a perfect alignment. The possibility of the hexagonal cell method of assembling a unit of a thin bundle is that it is possible to manufacture an SSC-type wire with 2.5 μm filaments in a complete array in a relatively short time and in an economical manner.

The initial shape of the circular cross section of a single core filament rod may be processed to be a hexagonal or circular cross section NbTi filament or filament of other material by selecting appropriate process condition variables. I understand.

While the particular embodiments of the invention have been described, it will be apparent to those skilled in the art that changes and modifications can be made in the broader aspect of the invention without departing from the invention. Let's do it. Therefore, the purpose of the appended claims is to cover all such changes and modifications within the true spirit and scope of the invention. The foregoing description and drawings are provided by way of illustration only and should not be construed as limiting the invention in any way. When considered in the proper context of the prior art, the actual scope of the invention is defined in the following claims.

[Brief description of drawings]

FIG. 1 is a schematic view of a typical triple extrusion method used in a conventional method for producing an ultrafine multifilamentary superconducting wire.

FIG. 2 is a schematic view of the hexagonal cell extrusion method of the present invention.

FIG. 3 shows a single NbTi rod coated with Cu and having a barrier layer.

FIG. 4 shows the billet assembly with a number of thin walled hexagonal tubes with the rod of FIG. 3 inserted into the outer extrusion case.

FIG. 5 is a cross-sectional view of a wire rod of a final product after being drawn to have a suitable size.

FIG. 6 is a view showing a billet assembly including Nb rods for manufacturing a final product of Nb ₃ Sn filament wire rods.

FIG. 7 is a part filled with a plurality of single-core filaments,
It is the figure which showed the hexagonal cell with which the copper material was filled in the remaining part.

FIG. 8 is a cross-sectional view of a metal container filled with hexagonal cells and single-core wire rods arranged around the cells.

[Explanation of symbols]

10 Single core filament rod 12 Barrier layer 14 Cu coating 16 thin wall hex tube 18 Extrusion container

Claims (10)

[Claims] 1. A method for producing a superconducting wire, comprising producing a plurality of filament rods having a circular cross section, arranging the filament rods having the circular cross section in a hexagonal close-packed pattern in a thin hexagonal tube, Multiple thin hexagons 2 packed with circular cross section filament rods
Forming a secondary element member, placing the thin-walled hexagonal secondary element member inside an extrusion container, and compressing the extrusion container thus assembled into a composite and extruding the composite to the desired final dimensions. A method characterized by. 2. The method for producing a superconducting wire according to claim 1, wherein the filament rod is a single core filament rod. 3. The method for manufacturing a superconducting wire according to claim 2, wherein the single core filament rod having a circular cross section has a diameter of about 2 to 3 μm. 4. The method for producing a superconducting wire according to claim 2, wherein about 3000 to 100,000 single filament filament rods having a circular cross section are used. 5. The method of manufacturing a superconducting wire according to claim 2, wherein the thin hexagonal tube has a thickness of about 50 to 500.
A method characterized by being used. 6. The method for producing a superconducting wire according to claim 2, wherein the single core filament is an Nb filament, an NbTi filament, an Nb alloy filament or an NbTi alloy filament covered with a barrier layer and a copper coating. A method characterized by the following. 7. The method for producing a superconducting wire according to claim 1, wherein the filament rod is a multifilamentary filament rod. 8. The method for manufacturing a superconducting wire according to claim 2, wherein about 6,000 to 50,000 single core filaments are incorporated into each of about 50 to 500 of the thin wall hexagonal tubes, and the hexagonal tube is extruded. A method characterized by being incorporated into a container, consolidated, extruded at a temperature of about 60 ° F to 1400 ° F, and drawn into a wire of the desired final dimensions. 9. The method for producing a superconducting wire according to claim 2, wherein the thin hexagonal tube is made of Cu, Ni or an alloy thereof. 10. A superconducting wire manufactured by the method according to claim 2.